As someone who collaborates with AI programming tools daily, I understand this feeling of exhaustion.
My project is a rich text editor SDK, with a considerable amount of code. AI tools are indeed very useful for writing glue code and generating boilerplate code. However, when faced with complex cross-browser compatibility issues (such as parsing mso-* CSS pasted from Word), AI is basically useless; I still have to understand the problem domain and write the code manually.
The idea of converting pseudocode to source code is interesting. The key is whether the pseudocode can clearly express the intent—if the intent itself is complex, the pseudocode may not be much shorter than the actual code.
Having worked on rich text editors for several years, I've developed a strong sense of the limitations of native HTML capabilities.
Popovers and dialogs are indeed useful; our project's color picker popup has been moved to a popover, which is much cleaner than manually managing z-index and closing it by clicking outside.
However, contenteditable is an exception. While it appears to be a native rich text editing capability of browsers, the implementation varies significantly across different browsers. The same operation will produce completely different HTML in Chrome, Firefox, and Safari. Not to mention mobile browsers with their various unpredictable input behaviors. This is why editors like ProseMirror and Wordgard don't trust contenteditable output; instead, they maintain their own document model and only use contenteditable as an input layer.
While native HTML capabilities are improving, for complex interactive scenarios, such as rich text editing and clipboard parsing, JavaScript is still essential.
My project is a rich text editor SDK, with a considerable amount of code. AI tools are indeed very useful for writing glue code and generating boilerplate code. However, when faced with complex cross-browser compatibility issues (such as parsing mso-* CSS pasted from Word), AI is basically useless; I still have to understand the problem domain and write the code manually.
The idea of converting pseudocode to source code is interesting. The key is whether the pseudocode can clearly express the intent—if the intent itself is complex, the pseudocode may not be much shorter than the actual code.